VoxEU Column COVID-19 Economic history

Coronavirus from the perspective of 17th century plague

Between 1563 and 1665, London experienced four plagues that each killed one fifth of the city’s inhabitants. This column uses 790,000 burial records to track the plagues that recurred across London (epidemics typically endured for six months). Possibly carried and spread by body lice, plague always originated in the poorest parishes; self-segregation by the affluent gradually halved their death rate compared with poorer Londoners. The population rebounded within two years, as new migrants arrived in the city “to fill dead men’s shoes”.

From the 1340s to the 1720s, Europeans lived in well-justified terror of plague. In the space of a century, London alone suffered four plagues (1563, 1603, 1625, and 1665) that each killed about one fifth of the city’s population, alongside lesser plagues in 1593 and 1636.

In response to the ever present risk of devastating epidemics, the English government began in the late 16th century to compile weekly Bills of Mortality to record the number of deaths in each London parish, making specific note of those who had succumbed to plague. Though most of these early Bills were destroyed in the Great Fire of 1666, the burial records of nearly every London parish survive, in most cases going back to the 16th century. A small sample of these were used by Slack (1985) in his extraordinary study of London plague; in our study (Cummins et al. 2016) we analysed all 790,000 surviving records to see how plague spread week by week across the city’s parishes from 1560 to 1665.

Plague epidemics differed from the present coronavirus pandemic in two ways. (It does not need to be emphasised that we are not seeking in any way to downplay the gravity of the current outbreak by comparing it with the savage death rates of plague.) The first is that plague in the 17th century was almost invariably lethal. Every year, several ranchers and hikers in the western US still contract plague; although they are usually strong, fit people receiving careful medical attention, without antibiotics only one third would survive. The second way that plague differed from coronavirus is that plague spread extremely slowly.

Each plague followed a similar course. Deaths would start in May in the poorest parishes and slowly rise as the disease ground its way across the slums and shanty towns that surrounded the city, and then spread inside the walls. The weekly growth of the number of fatalities in each of the largest plagues is shown in Figure 1. Despite being a small, densely populated city where inhabitants from all parts of town jostled against each other in the narrow streets, deaths rose at a slow pace, on the order of 50% per week compared with the 33% per day associated with outbreaks of coronavirus.

Figure 1 Cumulative plague deaths in the four major epidemics

Source: Cummins, Kelly and O Grada (2016)

The second difference is the persistence of outbreaks: plague epidemics usually lasted over six months until they tapered away in the cold of winter. This reflects the fact than any sort of social distancing was impossible: all except the very rich faced a bleak choice between work as usual or starvation.

From the late 16th century on, rural England had a surprisingly effective system of income support in the form of the Poor Law, which prevented poor harvests leading to mass mortality as they had in the middle ages and still did across most of Europe – among the rich as well as the poor: remember that before 20th century advances in public health, famine killed people by infectious disease before they had a chance to succumb to hunger (Kelly and Ó Gráda 2013). London, however, had little in the way of poor relief; even if it had, that relief would have rapidly been overwhelmed by the needy.

Figure 2 Weekly plague deaths, 1655

Source: Cummins, Kelly and O Grada (2016)

The weekly spread of plague across the city is shown in Figure 3. In the top panel, each parish is shaded according to the intensity of deaths relative to the same weeks in previous years using a standard epidemiological surveillance system: colour begins to change as deaths cross an alarm threshold that warns of potential disease outbreaks. Parishes shaded dark red are suffering ten or more times their alarm threshold. Each major plague showed the same pattern: deaths would start to rise in the poorest parishes in early summer and would then move around the parishes and across the walls before taking over the entire city by late summer. Excess deaths would tail off towards October as the weather began to cool. In the bottom panel, the number of weekly deaths in each parish is shown by the size of the circles, with the intensity of plague relative to their usual death rate given by their brightness.

Figure 3 Weekly spread of plague in 1665

Source: Cummins, Kelly and O Grada (2016)

What is notable about Figure 3 is that the central parishes with walls suffered less devastation than elsewhere. London plague was not an equal opportunity killer and its lethality tracked the changing social geography of the city. In the 1580s, tax records show that affluent Londoners spread themselves fairly uniformly across the city. By the 1630s, when we next have data on the location of the well off, the rich were strongly concentrated in the central parishes inside the walls. This social segregation led to a marked fall in the mortality of these parishes. Figure 4 shows how the central parishes went from having the highest increase in deaths in 1563 (6.3 times their usual rate), to less than half the levels of the poorer Liberties outside the walls and the more distant out-parishes that were increasingly becoming shanty towns for the poor. To be sure, part of this decline in deaths reflected the fact that the wealthy could flee the city – we see this in the reduced number of births in these parishes – but most of it reflects lower death rates.

Figure 4 Burials in London districts relative to average of previous years

Source: Cummins, Kelly and O Grada (2016)

Although devastating, the impact of plague on London’s population was surprisingly transitory. Within two years of each visitation, population as measured by births had returned to its previous level, as migrants from the countryside flowed in “to fill dead men’s shoes”.

Despite their importance for the urban development of Europe between the 14th and 18th centuries, plague outbreaks after the Black Death have received surprisingly little attention. Alfani and Murphy (2016) provide a useful recent survey of the literature, and Alfani and Bonetti (2019) give a detailed analysis of an outbreak in one Italian town in 1630, finding that women and men were equally likely to die and that mortality risk peaked between ages 40–60.  One of us (Cummins 2017) studied plague mortality amongst Europe’s elites from 1346 to 1666; among nobles, women died at higher rates than men. The classic study of London plague is Slack (1985).

Plague remains one of the least understood diseases, reflected in the saying among epidemiologists that “if you think you understand plague then you know very little about plague”. Particular puzzles include: why plague seems to have disappeared between the First, Second, and Third Pandemics; how it can remain quiescent for decades before suddenly flaring up again; and why some outbreaks were so much deadlier than others. Indeed, the low death rates of plague since it reappeared in the late 19th century compared with the savage mortality of earlier centuries led many to question whether earlier plagues were even caused by modern Yersinia pestis (this was resolved by the discovery of plague DNA in the dental pulp of corpses in medieval plague pits).

Rats are central to the historiography of plagues. Yet, earlier plagues left no accounts of the mounds of dying rats that are a feature of later outbreaks, and the Black Death devastated rat-free Iceland. The idea that plague can be transmitted directly by human fleas and body lice, initially developed by Drancourt, Houhamdi, and Raoult (2006), is now widely accepted. The importance of body lice is further supported by 19th century accounts noting that outbreaks of plague were usually accompanied by typhus (a fever spread by lice), and that typhus often showed the same symptoms as plague: swollen lymph glands and intradermal bleeding (“ring of roses”), followed by swift, agonising death.

Western Europe experienced its last and utterly apocalyptic plague outbreak in Marseilles in 1720. When did plague disappear from London? The London Bills last mention plague deaths in 1679. But the seasonality pattern associated with plague, with deaths peaking in September, persisted for several decades more. The modern pattern, whereby deaths peak in late winter, only starts in London in the 1730s. Until then, deaths still showed the distinctive pattern of the plague era, especially in poorer parishes, suggesting that plague may have endured at low levels until then.


Alfani, G and M Bonetti (2019), “A survival analysis of the last great European plagues: The case of Nonantola (Northern Italy) in 1630”, Population Studies 73:101–118.

Alfani, G and T E Murphy (2016), “Plague and Lethal Epidemics in the Pre-Industrial World”, Journal of Economic History 77:314–343.

Cummins, N, M Kelly and C Ó Gráda (2016), “Living Standards and Plague in London, 1560-1665”, Economic History Review 69:3–34.

Cummins, N (2017), “Lifespans of the European Elite, 800–1800”, Journal of Economic History 77[2]: 406--439

Drancourt, M, L Houhamdi and D Raoult (2006), “Yersinia pestis as a Telluric, Human Ectoparasite-borne Organism”, Lancet Infectious Diseases 6:234–241.

Kelly, M and C Ó Gráda (2013), “Living Standards and Mortality in England since the Middle Ages”, Economic History Review 67[2]: 358-81.

Slack, P (1985), The Impact of Plague in Tudor and Stuart England, London: Routledge & Kegan Paul.

8,504 Reads